Phosphite (phosphorous acid): Fungicide, fertilizer or bio-stimulator?

Abstract

Phosphite (; Phi), a reduced form of phosphate (; Pi), is widely marketed as either a fungicide or fertilizer or sometimes as a biostimulant. This is confusing for both distributors and growers. The present paper explores data from various studies to clarify that Phi does not provide plant P nutrition and thus cannot complement or substitute Pi at any rate. In addition, Phi itself does not have any beneficial effect on the growth of healthy plants, regardless of whether it is applied alone or in combination with Pi at different ratios or different rates. The effect of Phi on plants is not consistent, but is strongly dependent on the Pi status of the plants. In most cases, the deleterious effect of Phi is evident in Pi-starved, but not Pi-sufficient, plants. Plants fertilized with Pi allowing for approximately 80–90% of its maximum growth might still be at risk of the effect. This negative effect becomes more pronounced under more seriously Pi-deficient conditions. Although a number of studies have shown positive crop responses to Phi, these responses are likely to be attributable to the suppression of plant diseases by Phi and/or to Pi formed from oxidation of Phi by microbes. In addition, indirectly providing P by Phi-to-Pi oxidation is not an effective means of supplying P to plants compared with Pi fertilizer. An understanding of these issues will aid the right selection of fertilizer as well as minimize the harmful effects of Phi use on crops.

Key words:

• biostimulant
• fertilizer
• fungicide
• phosphate
• phosphite

INTRODUCTION

Phosphorus (P) is one of the major essential macronutrients required by all living organisms. In nature, P does not exist as a free element, but rather in combination with other elements, such as oxygen (O) or hydrogen (H). It occurs in a fully oxidized form (P has valence of +5) as phosphate anion (; Pi) and with one less oxygen (P has valence of +3) as phosphite (; Phi). The conjugate acid of the phosphite anion is phosphorous acid (H₃PO₃). In general, the term phosphite is commonly referred to as the salts of phosphorous acid and the term phosphonate is used to mean phosphite ester containing a carbon–phosphorus (C–P) bond that is chemically distinct from the labile carbon–oxygen–phosphorus (C–O–P) bond found in phosphate ester (White and Metcalf 2007).

For many years, Pi compounds were considered to be the only form that could supply P nutrition to plants. In fertilizer, P is normally found in the form of phosphoric acid (H₃PO₄) and its salts, such as, triple super phosphate, ammonium phosphate and potassium phosphate. All of these forms readily disassociate to release hydrogen phosphate () and dihydrogen phosphate (), which are used by plants. Phi compounds have been recognized as excellent fungicides for controlling many important plant diseases caused by Oomycetes, particularly Phytophthora sp. (Fenn and Coffey 1984; Foster et al. 1998; Grant et al. 1992; Guest and Grant 1991; Guest et al. 1995; Jackson et al. 2000; Jee et al. 2002; Smillie et al. 1989). However, they are not considered to be P fertilizers because an early study examining different P compounds by MacIntire et al. (1950) concluded that Phi compounds were a very poor source of P for crops compared with Pi fertilizers. Renewed interest in Phi as a plant nutrient has increased since the early 1990s when Lovatt (1990a) discovered that P deficiency caused changes in nitrogen metabolism and that the foliar application of potassium phosphite to P-deficient citrus recovered the biological response of plants and consequently restored normal plant growth. Additional studies by Lovatt (1990b) further suggested that soil or foliar application of Phi could replace Pi as a source of P in avocado. A series of US patents, such as US Patent numbers 5514200 (Lovatt 1996), 5830255 (Lovatt 1998) and 5707418 (Hsu 1998) describe formulations containing Phi that are suitable as P fertilizers for plants. The work of Lovatt led to the first commercialization of Phi compounds as fertilizer. This fertilizer was sold under the trademark Nutri-phite (Biagro Western Sales, Visalia, CA, USA). Many new P fertilizers based on Phi are now being marketed. The list of Phi products that are available in the American and European markets that are sold as fertilizers now includes 10 different brand names (see Table 1) (Leymonie 2007). All of these products are formulated as alkali salts of phosphorous acid (H₃PO₃) and have been registered under the fertilizer laws. Some are even described more as “biostimulants” than as fertilizers. However, the representation and the use of Phi-containing products as sources of plant nutritional P have been subjected to ongoing controversy. This has also created much confusion for distributors and growers (Leymonie 2007). There are numerous publications indicating that Phi can be well absorbed by leaves and roots, but has no utility for plants as a P fertilizer (Carswell et al. 1996; Forster et al. 1998; Schroetter et al. 2006). Instead, Phi was found to have negative effects on the growth and metabolism of P-deficient plants by suppressing the typical molecular and developmental responses of plants to P deficiency (Abel et al. 2002; Carswell et al. 1996, 1997; Ticconi et al. 2001; Varadarajan et al. 2002). Phi intensifies the deleterious effects of P deficiency by tricking Pi-deprived plant cells into sensing that they are Pi sufficient, when in fact their cellular Pi content is extremely low (McDonald et al. 2001a).

Table 1 From fungicides to fertilizers: The marketing of some products with phosphorous and phosphite as the active ingredient

Product	Company	Country	Active ingredient	Marketed as
Aliette	Bayer Cropscience	Germany	Aluminum phosphite	Fungicide
Nutri-phite	Biagro Westem Sales	USA	Phosphite and organic acids	Fertilizer
Ele-Max	Helena Chemical	USA	Phosphorous acid	Foliar fertilizer
ProPhyt	Luxembourg-pamol	USA	Monopotassium phosphite	Systemic fungicide
Nutrol	Lidochem	USA	Potassium phosphite	Fertilizer and fungicide
Phostrol	NuFarm America	USA	Phosphorous acid	Biochemical pesticide
Agrifos	Liquid Fert Pty (Agrichem)	USA	Monopotassium phosphite	Fungicide
Foli-r-fos 400	UiM Agrochemicals	Australia	Monopotassium phosphite	Fungicide
Fosphite	Jh Biotech	USA	Monopotassium phosphite	Fungicide
Lexx-a-phos	Foliar Nutrients Inc	USA	Monopotassium phosphite	Fungicide
Trafos line	Tradecorp	Spain	Potassium phosphite	Fertilizer and defense stimulator
Phytos’K	Valagro	Italy	Potassium phosphite	Biostimulant (registered as EC fertilizer)
Phosfik line	Biolchem	Italy	Phosphorous acid	EC fertilizer
Fosfisan, Vigorsan	Agrofill	Italy	Potassium phosphite	Defense stimulator (registered as fertilizer)
Geros-K	L-Gobbi	Italy	Potassium phosphite	EC fertilizer
Kalium Plus	Lebosol	Germany	Potassium phosphite	EC fertilizer
Frutoguard	Spiess Urania	Germany	Potassium phosphite	EC fertilizer
Foliaphos	Plantin	France	Potassium phosphite	EC fertilizer
Source: Leymonie 2007. EC, European Commission (for identifying chemicals).

The confusion about Phi became greater when some scientists (Lovatt and Mikkelsen 2006; Watanabe 2005) claimed that the negative effects of Phi on plant growth observed in many studies resulted from the inappropriate use of this material, for example, as a primary source of P or in excessive amounts. Lovatt and Mikkelsen (2006) emphasized that “since phosphite is chemically different from phosphate, these differences must be taken into consideration to avoid plant toxicity” and that Phi, if used at appropriate rates, can provide stimulation to plants that may not occur with Pi. Lovatt and Mikkelsen (2006) suggested that Phi is more than just a fungicide; for example, it increases floral intensity, yield, fruit size and total soluble solids. In addition, combinations of Phi and Pi ions are believed to be more effective than either Pi or Phi alone in plant assimilation (Foster et al. 1998; Young 2004).

What are the true effects of Phi on the growth of plants aside from the fungicidal actions? Does Phi have any nutritional impact on plants? Can Phi provide stimulating effects to healthy plants? Can a combination of Phi and Pi be more effective in plant assimilation than either ion alone? This article will explore data from various studies to shed light on these questions.

PHOSPHITE IS NOT A FERTILIZER AND DOES NOT HAVE ANY BENEFICIAL EFFECT ON THE GROWTH OF HEALTHY PLANTS

Aside from the early works of Lovatt (1990a,b) demonstrating that Phi was readily taken up through leaves and could replace Pi as a source of P in the metabolism of citrus and avocado, a number of studies have highlighted the potential use of Phi as P fertilizer for plants. Albrigo (1999) demonstrated that foliar application of potassium phosphite on Valencia orange trees in Florida clearly increased both the yield (boxes per hectare) and quality (orange juice soluble solids per hectare) over untreated controls, although the mechanism of the Phi effect remains unclear. Another study by Lovatt (1999) on Navel orange trees indicated that foliar application of potassium phosphite in May and July significantly increased the yield, fruit size and total soluble solids. The improvement in the yield and fruit quality after foliar application of potassium phosphite was interpreted as a response of citrus fruit to increased P nutrition. Similarly, foliar application of potassium phosphite has been reported to accelerate the flowers and fruit setting and, hence, significantly increase the yield of Satsuma orange compared with untreated controls or a foliar Pi treatment (Watanabe 2005). Rickard (2000) summarized studies on crop responses to commercial Phi-derived P fertilizers. Most of the studies in his review were carried out under field conditions and all results showed that soil or foliar application of Phi fertilizer consistently improved the yield and quality of many crops, such as celery, onion, potatoes, peaches, orange and cotton. However, there is no evidence that Phi can be used directly by plants as a source of nutritional P. In addition, data showing that the effectiveness of Phi-derived P fertilizer is equal to or better than that of conventional Pi fertilizers are very rare.

In contrast, numerous other studies have indicated that Phi compounds cannot be used as P fertilizer by plants. Hydroponically cultivated tomato and pepper plants treated with either commercial Phi or technical Phi (prepared from acid phosphorous and neutralized with KOH) exhibited a significant reduction in growth compared with Pi-fertilized plants (Forster et al. 1998; Varadarajan et al. 2002). A study by Schroetter et al. (2006) on maize plants indicated that foliar application of potassium phosphite did not improve the growth of maize plants in a field trial under either Pi-deficient or Pi-sufficient conditions. In his pot experiment, the growth of maize plants treated with potassium phosphite as the sole P source via either soil or foliar applications was strongly inhibited. The negative effects of Phi ranged from stunted growth to complete death. A negative effect of Phi was also found in Brassica nigra seedlings grown in vitro (Carswell et al. 1996), in Brassica napus cell suspension (Singh et al. 2003), and in Ulva lactuca culture (Lee et al. 2005).

Although most plants readily absorb and translocate Phi, it does not appear to be readily oxidized or metabolized in plants (Carswell et al. 1996, 1997; Guest and Grant 1991). Instead, Phi is found to be deleterious to Pi-starved, but not Pi-fertilized plants by suppressing a wide range of the plant's responses to Pi deficiency (Carswell et al. 1996, 1997; Ticconi et al. 2001; Varadarajan et al. 2002), consequently exacerbating the deleterious effects of Pi starvation (Mcdoldnan et al. 2001a). Enhanced root growth or an increased root to shoot ratio, the hallmark of Pi stress responses, were found to be strongly inhibited by Phi in B. nigra (Carswell et al. 1996), tomato (Varadarajan et al. 2002), spinach, komatsuna and celery (Thao et al. 2008a,b; Thao and Yamakawa 2008). Pi-starvation-induced root growth (root hairs, root length and root density) in Arabidopsis was also significantly reduced by Phi treatment (Ticconi et al. 2001). In addition, Phi has been shown to prevent the acclimation of plants and yeast to Pi deficiency by specifically suppressing the expression of Pi-starvation-inducible genes (Carswell et al. 1996, 1997; Mcdoldnan et al. 2001b; Ticconi et al. 2001; Varadarajan et al. 2002). Biochemical adaptations to Pi starvation include increased synthesis of anthocyanins, presumably to adjust photosynthesis light reactions to the Pi-dependent Calvin cycle, and increased synthesis of enzymes for scavenging intra-cellular and extra-cellular P (Ticconi et al. 2001). In Arabidopsis, the accumulation of anthocyanins and the activities of Pi-starvation-inducible nucleolytic enzymes (ribonuclease, phosphodiesterase and acid phosphatases) were effectively prevented by Phi (Ticconi et al. 2001). Studies on B. nigra and B. napus by Carswell et al. (1996, 1997) showed that the induction of acid phosphatase, phosphoenolpyruvate phosphatase, inorganic pyrophosphate-dependent phosphofructokinase and high affinity plasmalemma Pi translocator by Pi limitation was strongly inhibited in the presence of Phi. Similarly, Varadarajan et al. (2002) found that the expression of Pi-starvation-inducible genes, such as LePT1 and LePT2 (high-affinity Pi transporters), LePS2 (acid phosphatase) and LePS3 and TPSI1 (novel genes) in Pi-starved tomato was greatly suppressed by Phi treatment. Examination of the Phi effect on two yeast pho mutants revealed that Phi targets PHO84, a high-affinity Pi transporter and putative component of a Pi-sensor complex (McDoldnan et al. 2001b). The suppression of Pi-starvation responses by Phi was selective and was not caused by any general cellular toxicity of Phi (Albel et al. 2002) because Phi did not affect the expression of auxin-inducible genes (Ticconi et al. 2001), total chlorophyll, the protein content or the activities of enzymes that are not associated with the Pi-starvation responses of plants (Plaxton and Carswell 1999). Data from yeast and plants suggest that Phi mimics Pi in signaling pathways, thereby suppressing various Pi-starvation-inducible responses and consequently intensifying the deleterious effects of Pi starvation.

Although numerous studies have highlighted the deleterious effects of Phi on plant growth, some researchers have claimed that these negative effects of Phi resulted from the inappropriate use of this material, such as the use of Phi as a sole P source or in excessive amounts (Lovatt and Mikkelsen 2006; Watanabe 2005), to further dissect the Pi-starvation responses of plants. These researchers suggested that Phi, if used at appropriate rates, can provide stimulation to plants that might not occur with Pi. Furthermore, a combination of Phi and Pi ions is believed to be more effective than either ion alone in plant assimilation (Young 2004). Forster et al. (1998) found that Phi did not perform well as a Pi fertilizer, but they did observe growth enhancement of tomato plants treated with a mixture of Pi and Phi when compared to Pi alone.

Recently, a series of studies by Thao et al. (2008a,b; Thao and Yamakawa 2008 and H. T. B. Thao et al., unpubl. data, 2008) on various crops (spinach, komatsuna, celery and lettuce) using different methods and rates of Phi application was conducted to intensively investigate the effects of Phi as well as a combination of Phi and Pi on the growth and P nutrition of plants. All of these studies were short-time experiments (approximately 1 month) conducted in a greenhouse and many of them used water culture to minimize the interfering effects of pathogens as well as Phi to Pi conversion, so that the true effects of Phi on plant growth could be assessed. The authors did not detect any beneficial effect of Phi on plant growth. When Phi was applied to the roots in combination with Pi at different Pi : Phi ratios, for the same total amount of P applied to the roots at either low or high total P levels, the growth of both spinach and komatsuna drastically decreased as the proportion of Phi increased (Thao et al. 2008a,b). These results suggested that Phi was not a substitute for Pi at any rate, and that there was no stimulating effect from any Phi–Pi combination on plant growth. Foliar application of Phi at a rate (0.05% P₂O₅) that ensured no damage to the plant leaves significantly increased the P and Phi contents in the plant tissues, but did not improve plant growth, and the growth of komatsuna actually decreased. In hydroponically cultivated celery, the addition of Phi at levels ranging from 0.1 to 2 mmol L⁻¹ into the nutrient solution did not improve plant growth under either low (0.05 mmol L⁻¹) or high (0.5 mmol L⁻¹) Pi supply (Thao and Yamakawa 2008), and the growth of low Pi-fertilized celery was significantly reduced by 2 mmol L⁻¹ Phi. In lettuce (H. T. B. Thao et al., unpubl. data, 2008) the addition of Phi to the nutrient solution at different rates ranging from very low (0.05 mmol L⁻¹) to relatively high (2 mmol L⁻¹) substantially increased the total P and Phi in both shoots and roots, but did not improve the plant growth under various Pi supplied levels (0.05, 0.1, 0.15 and 0.3 mmol L⁻¹ as Pi levels, respectively, for approximately 50, 80, 90 and 100% of the maximum plant growth in hydroponics) (Fig. 1). The results in lettuce further showed, regardless of the Pi level, that the addition of Phi at different rates did not have any stimulating effect on plant growth. This confirms that a combination of Phi and Pi ions is not more effective than either ion alone in plant assimilation, which is not as expected (Forster et al. 1998; Young 2004). Taken together, all these results consistently show that plants are incapable of directly using Phi as a P source and thus Phi cannot complement or substitute Pi fertilizer at any rate. Phi does not have any beneficial effect on the growth of healthy plants, regardless of whether it is applied alone or in combination with Pi at different ratios or rates. These results corroborate evidence that Phi is not really used in plants as a P fertilizer and that it has negative effects on the growth and metabolism of P-deficient plants (Carswell et al. 1996, 1997; Forster et al. 1998; Schroetter et al. 2006; Ticconi et al. 2001; Varadarajan et al. 2002).

Figure 1  Effects of phosphite (Phi) treatments on shoot fresh weight (FW) of hydroponic lettuce grown under different phosphate (Pi) supplies. The Pi levels 0.05, 0.1, 0.15 and 0.3 mmol L−1 were Pi supplies for 50, 80, 90 and 100% of maximum plant growth, respectively. Error bars are standard error (n = 3). ***P < 0.001. Different letters indicate significant differences between means within the same Pi supply by least significant difference tests (P < 0.05). (Source: H. T. B. Thao et al., unpubl. data, 2008).

WHAT CONTRIBUTES TO THE POSITIVE RESPONSES OF CROPS TO PHI APPLICATION IN SOME STUDIES?

The number of products based on Phi, such as potassium phosphites, magnesium phosphites and calcium phosphites, would not have expanded in the market if there was no benefit from the application of Phi. Why some studies have shown a negative effect of Phi on plant growth (Carswell et al. 1996, 1997; Forster et al. 1998; Schroetter et al. 2006; Thao et al. 2008a,b; Thao and Yamakawa 2008 and H. T. B. Thao et al., unpubl. data, 2008; Ticconi et al. 2001; Varadarajan et al. 2002) and others have found positive crop responses to Phi (Albrigo 1999; Lovatt 1990a,b, 1999; Rickard 2000; Watanabe 2005) is not clear. It should be noted that although a number of studies have found positive crop responses to Phi, there is no evidence to date that can clearly demonstrate that plants use Phi directly as a P nutrient. We are also unaware of any plant enzyme that could oxidize Phi to Pi. To the best of our knowledge, there are no data showing a better response to Phi than to Pi of crops that are not the hosts of Oomycete pathogens. Furthermore, most studies showing positive crop responses to Phi were conducted in the field (Albrigo 1999; Lovatt 1990a,b, 1999; Rickard 2000; Watanabe 2005) where pathogens could interfere with plant growth (McDonald et al. 2001a). It is well known that Phi is able to effectively control many plant diseases caused by species of pseudofungi belonging to the order Oomycetes, particularly Phytophthora sp. (Fenn and Coffey 1984; Foster et al. 1998; Grant et al. 1992; Guest and Grant 1991; Guest et al. 1995; Jackson et al. 2000; Jee et al. 2002; Smillie et al. 1989). The fungicidal effects of Phi can act directly on the fungal pathogen and/or indirectly through stimulation of the plant defense response against pathogens (Jackson et al. 2000; Smillie et al. 1989). Thus, the benefits of Phi versus Pi in the field are likely to result from its fungicidal actions (McDonald et al. 2001a). In addition, it is now clear that various bacteria can metabolize Phi to Pi, for example, Escherichia coli, Pseudomonas stutzeri, Alcaligenes faecalis and Xanthobacter flavus (White and Metcalf 2007). Genes for the oxidation of Phi in diverse bacterial species have also been identified, such as Phn and PhoA (E. coli); Phn, htx and ptx (P. stutzeri) and htx-ptx (X. flavus). Although the process is slow and has no practical relevance (McDonald et al. 2001a), the slow release of orthophosphate by Phi-to-Pi conversion could contribute to the observed nutritional effects of Phi in field trials, particularly long-term trials. Thus, it is likely that the suppression of plant diseases by Phi and the indirect provision of Pi to plants after Phi-to-Pi oxidation by soil microbes are responsible for the beneficial effects of Phi on plants observed in some studies. In our recent studies (Thao et al. 2008a,b; Thao and Yamakawa 2008 and H. T. B. Thao et al., unpubl. data, 2008), under conditions to minimize the interfering effects of pathogens and Phi-to-Pi conversion, all results from various crops consistently showed that Phi does not provide plant P nutrition and does not have any beneficial effect on the growth of healthy plants, regardless of whether it is applied alone or in combination with Pi at different ratios or rates. Our results support the claim that Phi is relatively stable and is not oxidized or metabolized in plants (Carswell et al. 1996, 1997; Guest and Grant 1991).

In practical agricultural production, the application of Phi might have some positive effects on the growth and quality of plants as a result of the fungicidal properties of this material and some conversion of Phi into Pi by soil microorganisms. However, it should be kept in mind that Phi cannot be used directly by plants as a fertilizer and that Phi itself does not have any stimulating effects on the growth of healthy plants. Therefore, if the cropping environments do not have Oomycetes pathogens or if the crops are not hosts to this pathogenic group and conditions for Phi-to-Pi oxidation are unfavorable (e.g. hydroponic culture), one would expect the beneficial effect of Phi application to be negligible. Although Phi can be converted to Pi in the soil by microorganisms, this process is too slow to be agriculturally relevant compared with Pi fertilizers. The approximate half-life for Phi oxidation to Pi in soil is approximately 12–16 weeks (Adams and Conrad 1953). Furthermore, although some microbes are capable of oxidizing Phi, they preferentially use Pi over Phi as a source of P. The inoculation test of soil bacteria in a mixed culture of Phi and Pi by Adams and Conrad (1953) revealed that Phi was not used by the soil bacteria until all of the Pi in the culture was depleted and thus all traces of Pi would have been scavenged by the microbes before Phi oxidation occurred. In a study by MacIntire (1950) to evaluate the efficacy of different P compounds as P fertilizers for various crops in different soils, the application of Phi compounds was found to be very ineffective compared with Pi fertilizers. Phi was toxic to all first crops and only beneficial to the subsequent crops, highlighting the slow conversion of Phi to Pi. Recently, in 2004, farmers in South-Eastern Alabama, Southern Georgia and Northern Florida experienced toxicity problems in maize that appeared to be related to the use of ammonium phosphite, a non-conventional P fertilizer, as a starter fertilizer (Mitchell and Adams 2004). Therefore, the use of Phi as a P fertilizer is an inefficient means of supplying P to plants compared with Pi fertilizer (Lanschoot and Cook 2005; McDonald et al. 2001a).

THE PHYTOXICITY OF PHI DEPENDS ON THE PI STATUS OF THE PLANTS

A number of studies have emphasized the deleterious effects of Phi on Pi-deprived, but not Pi-fertilized plants (Carswell et al. 1996, 1997; Forster et al. 1998; Schroetter et al. 2006; Ticconi et al. 2001; Varadarajan et al. 2002). A recent study (H. T. B. Thao et al., unpubl. data, 2008) focusing on the effect of Phi in relation to Pi supply on hydroponic lettuce found that plants fertilized with Pi allowing for approximately 80–90% of its maximum growth were still vulnerable to Phi added to the nutrient solution at concentrations ranging from 0.2 to 2 mmol L⁻¹ (Fig. 1). This Phi range is comparable to the dose (100 p.p.m. H₃PO₃ equivalent to 1.2 mmol L⁻¹ Phi) recommended as a supplement into the nutrient solution for controlling Phytophthora root rot of lettuce in hydroponics (Jee et al. 2002). The severity of the Phi effect was dependent on both the Phi level and the Pi status of the plants. For example, lettuce fertilized with Pi for approximately 80% of its maximum growth or lower was still harmed by a low Phi concentration of 0.2 mmol L⁻¹, whereas lettuce fertilized with Pi for approximately 90% of its maximum growth was only negatively affected by Phi at a high rate (2 mmol L⁻¹); and under sufficient Pi supply (0.3 mmol L⁻¹) the addition of Phi up to 2 mmol L⁻¹ did not influence plant growth (Fig. 1). This result is consistent with our previous research on komatsuna and celery (Thao and Yamakawa 2008; Thao et al. 2008b), which suggested that the effects of Phi were highly dependent on the Pi status of the plants. Although plants differ in their sensitivity to Phi, the strong Pi-dependent effect of Phi is believed to occur in most plants. Understanding this can help to avoid the harmful effect of Phi-containing products to crops.

CONCLUSIONS

Our study has lead to the following conclusions:

1.	Although Phi can be absorbed by most plants through the leaves and/or roots, it cannot be used directly as a nutrient source and therefore cannot complement or substitute Pi fertilizer at any rate.
2.	Phi itself does not have any stimulating effects on the growth of healthy plants and Phi and Pi in combination do not provide any stimulating effects compared with Pi alone. Thus, if the cropping environment is unfavorable for Phi-to-Pi conversion and pathogens belonging to the Oomecetes group are not a problem, one would expect no beneficial effect of Phi on the crops.
3.	The effects of Phi on crops are strongly dependent on the P nutrient status of the plants. A deleterious effect of Phi was not evident in Pi-sufficient plants, but plants fertilized with Pi sufficient for approximately 80–90% of their maximum growth may still be at risk of the effect. The negative effect of Phi becomes more pronounced under more seriously Pi-deficient conditions. Therefore, Phi should not be applied to plants in sub-optimal Pi

This paper provides a clearer understanding of the effects of Phi on the growth and P nutrition of plants and will aid the selection of appropriate fertilizers as well as minimize the harmful effects of the use of Phi on crops. Our aim was to help reduce some of the confusion experienced by growers and distributors (Leymonie 2007) with regard to Phi.